select

Command select selects best gene models.

usage

$ ingenannot -v 2 select file.fof selected_genes.gff --noaed --clustranded --penalty_overflow 0.25 --use_ev_lg  --no_cds_overlap

positional arguments:

FoF	File of files, <GFF/GTF>TAB<source>
Output	Output Annotation file in GFF file format

optional arguments:

-h, –help	show this help message and exit
–noaed	If set, use precompute aed info available in gff file, no aed computation
–clutype CLUTYPE	Feature type used to clusterize: [gene, cds], default=cds
–clustranded	Same strand orientation required to cluster features, default=False
–evtr EVTR	Gff file of transcript evidence
–evpr EVPR	Gff file protein evidence, compressed and indexed with tabix
–evtr_source EVTR_SOURCE	Specify source for Gff file transcript evidence ex “stringtie”, default=undefined
–evpr_source EVPR_SOURCE	Specify source for Gff file protein evidence ex “blastx, miniprot”, default=undefined
–evtrstranded	Same strand orientation required to consider match with evidence, default=False
–evprstranded	Same strand orientation required to consider match with evidence, default=False
–penalty_overflow PENALTY_OVERFLOW	In case of a CDS is longer or violate constraint of intron with the tr evidence, add a penalty to the computation of the aed score range[0.0-1.0], default=0.0, no penalty
–nbsrc_filter NBSRC_FILTER	Number of sources required to bypass aedtr and aedpr filtering, default=max number of source + 1
–aedtr AEDTR	Minimum aedtr required when filtering default=1.0,
–aedpr AEDPR	Minimum aedpr required when filtering default=1.0,
–aed_tr_cds_only	For transscripts (short-reads and longreads), compute aed on CDS only, instead of Exon and CDS, with best score selection, default=False
–use_ev_lg	Use aed of long-read instead of aed_tr if better, default=False
–nbsrc_absolute NBSRC_ABSOLUTE	Number of sources required to keep a gene, default=1
–min_cds_len MIN_CDS_LEN	Minimum CDS len required default=90, prot with 30AA
–no_partial	In case of the best selected CDS is partial (no CDS nor STOP codon), export another CDS if possible, default=False
–genome GENOME	Genome in fasta format required with no_partial
–longreads LONGREADS	Gff file longread based transcript evidence, compressed and indexed with tabix
–longreads_source LONGREADS_SOURCE	Specify source for Gff file longread based evidence ex “Iso-Seq”, default=undefined
–longreads_penalty_overflow LONGREADS_PENALTY_OVERFLOW	In case of a CDS is longer or violate constraint of intron with the longread based transcript evidence, add a penalty to the computation of the aed score range[0.0-1.0], default=0.25
–gaeval	expect gaeval tsv file for each annotation file, File of files: <GFF/GTF>TAB<source>TAB<gaeval>
–prefix PREFIX	prefix for gene name, default=G
–no_export	export gff file containing the un-exported best genes
–no_cds_overlap	post process filter to remove worst CDS if overlapping with other CDS

inputs

File of Files (FoF) with all files to analyze. One per line such: <GFF/GTF>TAB<source>. If –gaeval option set, expect gaeval tsv file for each annotation file, File of files: <GFF/GTF>TAB<source>TAB<gaeval>. It is recommended to use option –no_cds_overlap to avoid overlapping CDS on same or opposite strand, especially in case you performed gene prediction on specific strand.

outputs

The main outtput is the gff file with selected genes. Many tags are added on mRNA features as AED scores:

chr_1        ingenannot      gene    615870  616337  .       -       .       ID=G_00002;gene_source=gene:ZtritIPO323_04g00197;source=RRES;
chr_1        ingenannot      mRNA    615870  616337  .       -       .       ID=G_00002.1;transcript_source=mRNA:ZtritIPO323_04t00197;source=RRES;Parent=G_00002;ev_tr=None;aed_ev_tr=1.0000;ev_tr_penalty=undef;ev_pr=None;aed_ev_pr=1.0000;ev_lg=None;aed_ev_lg=1.0000;ev_lg_penalty=undef;
chr_1        ingenannot      exon    615870  616337  .       -       .       ID=exon:G_00002.1;Parent=G_00002.1;
chr_1        ingenannot      CDS     615870  616337  .       -       0       ID=cds:G_00002.1;Parent=G_00002.1;
chr_1        ingenannot      gene    617267  619263  .       -       .       ID=G_00003;gene_source=gene:gene:chr_1g0002351;source=eugene;
chr_1        ingenannot      mRNA    617267  619263  .       -       .       ID=G_00003.1;transcript_source=mRNA:mRNA:chr_1g0002351;source=eugene;Parent=G_00003;ev_tr=SCA3419A35.107.1;aed_ev_tr=0.0920;ev_tr_penalty=no;ev_pr=None;aed_ev_pr=1.0000;ev_lg=PB.103.1;aed_ev_lg=0.0000;ev_lg_penalty=no;
chr_1        ingenannot      exon    617267  617497  .       -       .       ID=exon:G_00003.1;Parent=G_00003.1;
chr_1        ingenannot      exon    617552  619263  .       -       .       ID=exon:G_00003.2;Parent=G_00003.1;
chr_1        ingenannot      CDS     617367  617497  .       -       2       ID=cds:G_00003.1;Parent=G_00003.1;
chr_1        ingenannot      CDS     617552  619004  .       -       0       ID=cds:G_00003.1;Parent=G_00003.1;

If the option –no_export was set, best filtered gene models (aed filter, nb source, cds_overlap) are exported in a separate gff file.